DESCRIPTION: Combined prevention and care for HIV among persons who inject drugs (PWID) has led to dramatic reductions in injecting-related HIV transmission in a number of locations, including Holland, France, Vancouver, and numerous cities in the United States. Simultaneous implementation of multiple interventions has generated synergistic effects much greater than the effects of individual interventions. This proposed research would apply the same logic of implementation of multiple evidence-based interventions to initiation into injecting drug use. Reducing initiation into injecting drug use would not only reduce transmission of blood-borne viruses (HIV and HCV) but also the many other individual and societal problems associated with injecting drug use. A broad conceptualization of initiation into injecting drug use will be utilized, from first injection to the transition process through which injecting becomes th dominant mode of drug administration. There are a number of interventions with demonstrated effectiveness in reducing initiation into drug injecting, including: low threshold treatment for drg use, Break the Cycle (in which current injectors are taught to avoid modeling for potential new injectors), the Heroin Sniffer project (in which non-injecting drug users are taught skills and provided with social support to avoid injecting) and Couples-Based HIV prevention (which addresses drug use issues within sexual partnerships). The latter 3 interventions all include social network/social support components and are capable of changing social norms regarding injecting drug use. The research would be conducted in two sites: Tallinn, Estonia, which has very high rates of HIV infection among new drug injectors, and New York City, USA, where there is a growing problem of prescription opiate users transitioning to heroin injecting. Both sites have the organizational capacity for integrating new interventions with existing services, and have well- established, highly productive collaborations among researchers, public health officials and community based service providers. Cohort studies would be used to assess willingness to participate and the outcomes of individuals participating in the multiple interventions. Mathematical modeling would then be used to address scale up and the synergistic effects of combined implementation. The research is ideally suited to the Avant Garde format, and the investigator has a long history of innovative research in HIV among drug users.